Good article on why you can't compare dyno #s

[Z K]

Just saw this. Informative article.

The moral of the story is that trying to compare dyno numbers from different dynos, locations, and weather conditions for bragging rights is borderline idiotic. The best case for trying to compare chassis dyno numbers between different cars is if both cars were tested on the same dyno in similar weather conditions. Use the dyno as intended, as a tool to measure how vehicle setup changes alter power output. Don't bench race.

http://www.motoiq.com/magazine_artic...lieve-one.aspx

[J08M3]

not very informative, just says dynos can't be compared. but thanks for link

[Z K]

Originally Posted by J08M3

not very informative, just says dynos can't be compared. but thanks for link

It goes into the reasons why, variables and calculations. I doubt you'd read the whole article in this short time period.

Something as simple as dyno strap tension can change the dyno reading.

[J08M3]

Originally Posted by Z K

It goes into the reasons why, variables and calculations. I doubt you'd read the whole article in this short time period.

Something as simple as dyno strap tension can change the dyno reading.

I read it, it's only 4 short paragraphs and a couple of tables. Nothing very in depth, just the usual stuff we always talk about here. Sorry, all I got from it.

[Z K]

Originally Posted by J08M3

I read it, it's only 4 short paragraphs and a couple of tables. Nothing very in depth, just the usual stuff we always talk about here. Sorry, all I got from it.

Great, glad you got the info.

[rjd598]

i think most people can come up with that conclusion. key is to keep an open mind to the various conditions of a dyno like the article said when comparing dyno runs

[schnell325]

The turbo Vespa on page 3 was cool!!

[OG_M3]

Originally Posted by J08M3

I read it, it's only 4 short paragraphs and a couple of tables. Nothing very in depth, just the usual stuff we always talk about here. Sorry, all I got from it.

you do know there are 3 pages right? think i counted more than 4 short paragraphs

[J08M3]

Originally Posted by OG_M3

you do know there are 3 pages right? think i counted more than 4 short paragraphs

Nope. Didn't see that guess that explains it.

[OG_M3]

Originally Posted by J08M3

Nope. Didn't see that guess that explains it.

all good, i missed it at first as well

[shay2nak]

+1 you can only compare dynos from the same shop on similar days. Same dynojets, different shops can give you different numbers and don't even try comparing dynojet to dynapack... dyno is basically tuning tool and to see what gains (+ or -) you've made with your mods. I'm sure you guys know this anyway.

[DLSJ5]

I think it is a little silly to compare different types of dyno's, DJ vs. DP or DJ vs. DD's, etc. in terms of making an argument that brand A makes more power than B, etc. But I think it's perfectly fine to compare Dynojet results, in the sports car enthusiast world the DJ really is the standard. For the most part, when the same CF's are applied to two different dynojet results, they are pretty accurate and unlike other dyno's they are very difficult to manipulate the results.

[Z K]

Originally Posted by DLSJ5

I think it is a little silly to compare different types of dyno's, DJ vs. DP or DJ vs. DD's, etc. in terms of making an argument that brand A makes more power than B, etc. But I think it's perfectly fine to compare Dynojet results, in the sports car enthusiast world the DJ really is the standard. For the most part, when the same CF's are applied to two different dynojet results, they are pretty accurate and unlike other dyno's they are very difficult to manipulate the results.

I wouldn't think the Dynojet is the "standard". They are just widely available and have been around longer so lots of places have it. Dynojets are not load bearing so the numbers you see are basically if there was no road resistance at all with the car essentially spinning the hubs freely. Dynojet numbers will consistently be higher than other dynos because of this. So if you want dyno #s for bragging rights, the dynojet is the one to use.

Mustang dynos are load bearing. Among tuners, it is one of the most widely used ones for accurate tuning as it will give closer real world road conditions. All the dyno operators in my area are running Mustang or Dyno Dynamics dynos.

Here's some more info.
http://g35driver.com/forums/266175-post1.html
Also Googling "Dynojet vs Mustang" will give a lot more info.

[E90///M3`]

i think this has been brought up and too many times already.

[DLSJ5]

Originally Posted by Z K

I wouldn't think the Dynojet is the "standard". They are just widely available and have been around longer so lots of places have it. Dynojets are not load bearing so the numbers you see are basically if there was no road resistance at all with the car essentially spinning the hubs freely. Dynojet numbers will consistently be higher than other dynos because of this. So if you want dyno #s for bragging rights, the dynojet is the one to use.

That is why, at least in the BMW platform, that is it considered the "standard." I didn't say it should be, didn't say it's the most accurate, nor did I say it was the best dyno for tuning, but it certainly is the most widely used / compared dyno in this platform and probably the world, especially in FWD or RWD platforms. In the GTR or AWD platforms the Mustang is probably the standard.

As far as a DJ reading high for bragging rights, most stock E9x M3's read 330-340whp SAE on a DJ, that's quite a bit of drivetrain loss to hit the rated 414BHP, but you're right, I think it's universally accepted and well known that DJ's for the most part do read higher than Mustang's.

[DLSJ5]

Originally Posted by E90///M3`

i think this has been brought up and too many times already.

Agreed.

[swamp2]

And this is all highly relevant for the debate on the topic of the effectiveness of M-DCT vs. MT in tuned M3s.

Apples to oranges baby, apples to oranges, just like I said (and said previously)...

[shay2nak]

Originally Posted by DLSJ5

I think it is a little silly to compare different types of dyno's, DJ vs. DP or DJ vs. DD's, etc. in terms of making an argument that brand A makes more power than B, etc. But I think it's perfectly fine to compare Dynojet results, in the sports car enthusiast world the DJ really is the standard. For the most part, when the same CF's are applied to two different dynojet results, they are pretty accurate and unlike other dyno's they are very difficult to manipulate the results.

you can't even compare dynojet results from different shops. Not unless you have a dyno baseline from a car who's power is widely known (using only NA). For example, an Elise will dyno 167/120 on a DJ. This is what it's supposed to show. There have been others who've pulled as low as 153/111 on DJs. Even the DJ that I use - I have a strong suspicion - is one of these lower reading DJs. I can't use my own car as a bearing because of it being supercharged and haven't been able to find an elise to do it (well i did and the guy's car was not stock at all when he said it was).

[DLSJ5]

Originally Posted by shay2nak

you can't even compare dynojet results from different shops. Not unless you have a dyno baseline from a car who's power is widely known (using only NA). For example, an Elise will dyno 167/120 on a DJ. This is what it's supposed to show. There have been others who've pulled as low as 153/111 on DJs. Even the DJ that I use - I have a strong suspicion - is one of these lower reading DJs. I can't use my own car as a bearing because of it being supercharged and haven't been able to find an elise to do it (well i did and the guy's car was not stock at all when he said it was).

Sure there can be anomalies, people that don't disclose mods, and of course the #'s are probably not going to be the same from DJ to DJ, but obviously most can see if there is something out of the ordinary when the #'s are vastly different.

[IMG]

The article makes some very valid and fairly obvious points:
1. Don't compare different brand dynos
2. SAE correction doesn't work when you run a dyno outside of the
formula's specified parameters.
3. It's possible to manipulate the results through strap tension
and altering the weather station data.

Luckily, nobody really does these things (other one or two known
dyno cheaters) and it's rare that a dyno is operated outside of
the SAE correction specification.

Here's a few things I also noticed about the article. The author
seems to have quite a bit of faith, in fact he even says SAE
correction is very accurate when used within it's operating
parameters. Another way to look at that is to say, that when you
don't compare apples to oranges (Mustang vs. Dynojet but keept
Dynojet vs. Dynojet) and when the dyno is run with SAE correction
within specifications, then the results really can be compared --
and compared pretty reliably (assuming strap tension is the
same).

Many people make a big deal about different weather conditions and
use that to say results can't be compared. But they don't realize
that SAE correction was designed specifically to correct for
weather and specificially to allow results to be compared. That's
literally why the SAE correction formula was invented. And
according to the author himself, when a dyno is run within the SAE
formula's specifications, then the correction is very accurate and
very reliable.

Originally Posted by swamp2

And this is all highly relevant for the debate on
the topic of the effectiveness of M-DCT vs. MT in tuned
M3s.

Swamp, I think you're trying to grasp at straws here to explain
why a supercharged DCT with less power can still trap higher than
a supercharged 6MT with more power. It sounds like you're
assuming that the dyno's were run incorrectly or that the weather
conditions played a major role in the outcome (visa vis dyno
operated outside of SAE specifications). As we've said a few
times already, some of the cars in those videos were dyno'd on the
same dyno with same exact weather conditions. Two cars were
dyno'd 24 hours apart on the same dyno with same exact weather
conditions, and two other cars were dyno'd on the same exact day,
on the same exact dyno only hours apart and also on the same exact
weather conditions. I've given you the links to the Dyno Database
files so you could download them yourself and verify the weather
conditions. It sounds like you've never done this because you're
still looking for anything to explain how these results fall
outside of your own understanding.

Originally Posted by swamp2

Please start another thread. A good topic might be
"the importance of DCT for drag racing" or something along those
lines. I promise I will have a closer look at your results and
reply with my own simulations.

One point does immediately stand out though. Your losses are
almost for sure incorrect. Incorrect based on real world
validations as well as text book values. Run your losses with a
standard power car and see how far off you are from the best
reported magazine times! If your losses are off (too high) but
results are right then you are right for the wrong reason and thus
a likely conclusion would that true power is actually a bit higher
than your input power.

I'll let you start the new thread. I already know how this plays
out in real life and how CarTest has already validated these
results itself. The only losses I find in CarTest are the
transmission losses. We've used the actual dyno charts and used
"At the wheels" for power rating. When you do this, CarTest
ignores the losses input into the program. Again, please verify
this yourself to make sure we are talking about the same thing.
It still looks like you're trying to find something to explain
these results because you don't understand it. Input the exact
values I posted for CarTest and then play with the losses. It
doesn't change the simulation results, and these simulation
results seem to perfectly match the real world results we've seen.
Don't forget, we've run some of these cars in real life --
including the ones that went to the same exact dyno on the same
exact day. The real life results were no different than the vBox
Racer and CarTest simulation results. So I'll leave it to you to
start a new thread if/when you find something wrong with the
simulations we've run, or if/when you find something wrong with
the dyno files I asked you to look at.

[shay2nak]

To make comparisons even more accurate, regardless of the weather, one could note the IAT during each pull. I typically have my OBD reader hooked up and I'll note it down.